The present application claims priority benefits under 35 U.S.C. xc2xa7119 to Russian Federation application N99126933/28 filed Dec. 17, 1999.
1. Field of the Invention
The present invention relates to the non-destructive testing of product quality, and in particular, to magnetic testing of elongated ferromagnetic objects like steel rods, tubes and wires to determine variances in a cross sectional area of object as well as the presence of local flaws.
2. Background of the Invention
A method and apparatus for magnetic non-destructive testing of elongated objects, e.g. steel ropes, by measurement of loss of metallic area (LMA) due to wearing and for local flaw (LF) detecting are described in U.S. Pat. No. 4,659,991, US C1. 324/241, Int.C1.G01N27/82, issuing in 1987. The method includes the axial magnetizing of the rope part by permanent magnets to a condition close to magnetic saturation and measuring the magnetic flux leakage variation near the rope surface by using sensing coils, in which an electromotive force (emf) is induced while rope moves relative to the magnets.
The disadvantage of the method of U.S. Pat. No. 4,659,991 is the insufficient accuracy of the LMA measurement as the method requires the measurement of a very small leakage flux change, resulting from a small value of the LMA (about one percent), at the significant initial level of the flux passing through the coils by the nominal rope cross sectional area.
The magnetic flux leakage change caused by LMA is quite commensurate with the flux leakage variation due to the magnetic flux instability of the magnet, particularly, when the environment temperature and hence, temperature of the permanent magnets, is changed. In addition, the emf in the coils is induced only while a rope moves relative to the coils and the emf depends on the rate of travel of the rope. Thus, the apparatus of U.S. Pat. No. 4,659,991 is relatively complicated and requires decreasing the speed and influence on measure readings. In addition, measurement is not possible when the rope is not traveling.
Also known in the prior art is a steel wire rope flaw detector in which there is no rope speed influence on readings (see, e.g. U.S. Pat. No. 5,565,771, Int.C1. G01N 27/72, 1996). The flaw detector disclosed in the ""771 patent contains magnetizing means with poles encircling a rope under test and Hall sensors. The sensors are located in gaps between poles of the magnetizing means and the rope. The second portion of the sensors are positioned in a gap of a magnetic core which is formed by a ferrous ring encircling the rope and by a ring-shaped magnetic shield with three-leg shape of cross section. The magnetic shield shields the second portion of the Hall sensors from magnetic field of the poles. The second Hall sensor portion provides LF detecting by magnetic flux leakage induced by the LF. A digital signal processor processes signals of all the Hall sensors.
The flaw detector of U.S. Pat. No. 5,565,771 is more convenient due to Hall sensor signals being independent of rope speed and their measure is possible when the rope is stopped. However, the flaw detector sensitivity to LF increases due to the ferrous magnetic core use.
The LMA measurement accuracy is often insufficient because of slight dependence of the magnetic flux density in the gaps between the poles of the magnetizing means and the rope from area of the rope cross section. The reason is that magnetic flux via the gaps is defined by the magnetizing means and by use of the high-energy modem permanent magnets. Magnetic flux density changes in the gaps around the rope, when the rope cross section changes only due to redistribution of magnetic flux portions: the main flux through the rope and the flux leakage in magnetizing means inter pole space. Relative change of the main flux is significantly less than the flux leakage relative change. That is why the relative change of Hall sensor voltage is insignificant. This results in an error caused by small voltage change of Hall sensor at high initial level of the voltage, and particularly, the temperature error is significant.
Further, LF detecting reliability of the flaw detector of U.S. Pat. No. 5,565,771 is often not sufficient. This appears to be due to the location of the Hall sensors in the magnetic core gap between the ring encircling the rope and the ring shield of three-leg shaped cross section. When the LF gets to the zone of sensitivity of the sensor, the leakage flux, induced by the LF and captured by the magnetic core, is divided into two portions. One portion goes through the gap with the Hall sensors and the remaining portion goes trough the three-leg shaped magnetic core bypassing the Hall sensors. Consequently, the LF signal of the Hall sensor decreases because of the bypass effect as like as LF detecting reliability. Accordingly, the accuracy of rope under test cross sectional area measurement is insufficient as well as the reliability of LF detecting.
Therefore the need exists for an apparatus for the magnetic testing of elongated ferromagnetic objects like steel rods, tubes, wires, wherein increased accuracy is achieved.
An object of the present invention is the solution of the problem of increasing the accuracy of cross sectional area measurements and local flaw detection in steel wire ropes. A further object of the invention is increasing the reliability of LF detecting in elongate ferrous objects.
In accordance with the method of the present invention, a section of an object under test, e.g. of a steel wire rope, is magnetized longitudinally by a magnetizing device having poles spaced along a longitudinal dimension of the wire rope. The longitudinal spacing of the poles along the longitudinal dimension of the rope defines a longitudinally extending inter-pole area between the poles. A magnetic field parameter, e.g. magnetic flux density, is measured by magneto-sensitive sensors forming a pair, in at least one pair of points in an area between the spaced poles (the inter-pole area) at the object surface. The pair of sensors is formed by two sensors placed along a line, which is parallel to the longitudinal axis of the rope in area of the most homogeneous magnetic field in the inter-pole space.
The rope cross sectional area is defined by a sum of the signals from the sensor pair. When the rope cross sectional area is changed, the magnetic fluxes through the rope and through the area surrounding the rope are redistributed. In particular, loss of the rope cross section metallic area (LMA) leads to an increase in magnetic flux leakage within the inter-pole space round the rope, which increases magnetic flux density in the inter-pole area. The relative increase of the leakage flux occurs significantly more than the relative decrease in the base flux. Therefore, a resulting signal change from the sensors within the inter-pole space is significantly more than the signal change of the sensors in the gaps between the poles and the rope. This provides an increased LMA measurement accuracy.
A further increase in accuracy is provided by a subtraction of additional magneto-sensitive sensor signals from the sum of the sensor pair signals. The additional sensor signal is provided from a sensor located in a gap between the rope under test and a pole of the magnetizing device. That is, along a radius extending from the longitudinal axis of the rope, there is the rope surface, the additional sensor and then the pole of the magnetizing device. The resulting second signal difference is used to measure LMA. Thus, an LMA measurement error, from the instability of magnetic flux density round the rope, e.g. from the influence of temperature variation on the magnetizing device, is decreased.
The presence of a LF in the rope is detected by a first signal difference of sensors forming the pairs. When a rope section under test contains no LF, then signals from the sensors in a pair equal each other and the first difference of the signals is close to zero. If the section, containing a LF, gets to a sensitivity zone of one of the sensors in the pair, then magnetic field homogeneity is destroyed in the zone due to a local magnetic flux leakage induced by the LF. The pair of the first sensors is thus unbalanced and the first signal difference differs from zero. Unavoidable radial displacements of the rope during a test do not destroy magnetic field homogeneity. Due to this, the output signal of the sensor pair remains close to zero, thus a noise level, e.g. because of rope vibration, is low. An increased signal to noise ratio is thereby provided as well as LF detecting reliability.
The flaw detector includes a magnetizing device, having poles spaced along the longitudinal axis of the rope and inducing a sufficient magnetic flux to provide magnetic saturation of the section of a rope being tested, a sensor unit located between the poles, a digital signal processor and a recorder. The sensor includes magneto-sensitive sensors and a magnetic core. The magnetic core is arranged as three ferrous elements surrounding the rope and positioned with respect to each other along the rope with gaps in which pairs of the magneto-sensitive sensors are located. Each the pairs is formed by two sensors located in gaps between the ferrous elements of the core on a direct line, which is parallel to the longitudinal axis of the rope.